The dose of therapeutic radiation that can be delivered to the brain is limited primarily by the risk of delayed brain necrosis which is debilitating and frequently fatal. Clinical and experimental dose-response data for this complication are limited. They are particularly inadequate for predicting how the brain will respond to dose fractionation in particular at doses of less than or equal to 2 Gy. Traditional isoeffect formulae are almost certainly inappropriate for the brain and newer formulations are suspect at low doses per fraction. Experimental studies on the response of brain to radiation have been limited to histology and lethality and no quantitative data are available on responses to clinically relevant radiation doses and fractionation schemes. We intend to measure the response of mouse brain to different single and multifractionated doses of radiation (down to 1 Gy per fraction) and to derive accurate isoeffect dose relationships. In addition, the effect of varying the interfraction interval on responses will be measured. "Remembered" dose will also be assessed as will the potential usefulness of radioprotectors. We will use sensitive immunological (ELISA) methods to measure radiation-induced changes in brain cell type-specific proteins in brain and body fluids. In this way we will measure responses by different brain cell populations and establish the kinetics of their damage and recovery. Alterations in brain cell-specific proteins as a function of time and dose will be corroborated by direct histological, immunohistochemical and autoradiographical examination. We believe that the isoeffect relationships that are established should serve as a guide in the establishment of new radiotherapeutic regimens and lead to better understanding of the biological bases for brain responses to radiation. In addition, in the future, the assays will be clinically applicable to the measurement of brain damage caused by radiation and other cytotoxic agents.